Implementation Suggestion

Hi All, how's it going, hope everyone's doing fine :)

I am just getting started with SVN, and still trying to figure out how to
use SVN to help with our deployment cycle

1. Currently we have development, staging and production environment

2. We will use a tool i.e. winmerge to compare between environments, and
deploy new codes from beta to staging and then from staging to production

3. Usually, one developer will go through this process with the developer
of the codes to qualify the code changes. This will result with clean new
codes to be deployed.

4. We call process 2 and 3 as synchronization process

5. Codes are written and saved in the development server rather than
developer's computer

6. Synchronization process has proved to be quite a lengthy process and if
we need to roll-back the codes, we have to manually source from the 'before
synch backup files' and know which files need to be replaced

7. All development, staging and production source codes are kept in folders
that reflects the URL of the system i.e.
https://application-a.com/*development*/index.html,
https://application-b.com/*staging*/index.html and,
https://application-a.com/*production*/index.html
rather than using the convention of trunk, branches, and tags

8. Is it possible to use svn to
I. Have *clean new codes* (please refer no. 3) ready on demand?
II. Roll-back codes to previous version automatically

9. I welcome any suggestion on how we can better implement this, if need
be, how we can do this with trunks, branches and tags?

Thank you very much and have pleasant day everyone
Received on 2015-01-31 06:56:44 CET